Abstract

Background Prostate adenocarcinoma (PRAD) is a highly prevalent malignant tumor in males and exhibits substantial heterogeneity. Identifying key associated genes is therefore critical for improving disease diagnosis and therapeutic strategies. Methods In this study, we first analyzed gene expression profiles to identify differentially expressed genes (DEGs) between PRAD and normal tissues. Subsequently, a weighted gene co-expression network analysis (WGCNA) was performed to elucidate the functional roles and signaling pathways of key genes, supported by Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and gene set enrichment analysis (GSEA) enrichment analyses. To prioritize robust diagnostic markers, three machine learning algorithms, least absolute shrinkage and selection operator (LASSO) regression, support vector machine with recursive feature elimination (SVM-RFE), and random forest, were integrated, and their intersection yielded six core genes: AOX1, APOBEC3C, C8orf88, GCNT4, NOX4, and PHYHIPL. Immune infiltration analysis via the CIBERSORT algorithm revealed that these core genes were significantly correlated with multiple immune cell populations. Notably, NOX4 was highly expressed in PRAD tissues, and its expression was negatively correlated with M2 macrophages while positively correlated with neutrophils. To investigate the functional role of NOX4, we constructed both NOX4 knockdown and overexpression models in the DU145 and PC3 PRAD cell lines, respectively. Cell Counting Kit-8 (CCK-8) and wound healing assays demonstrated that both NOX4 knockdown and overexpression markedly inhibited the proliferation and migration of PRAD cells. Results Results from the DCFH-DA fluorescence assay showed that NOX4 overexpression significantly elevated intracellular reactive oxygen species (ROS) levels and activated the PI3K/AKT/ERK signaling pathway in PRAD cells.